Controlling the spread of Acquired Immunodeficiency Syndrome (AIDS) is a major public health challenge. While currently concentrated in selected subgroups (homosexual/bisexual men, intravenous drug users), it is clear that all persons have the potential to develop the disease. At present, behavior change (specifically, the adoption of safer sexual practices) is the most effective means known for controlling the spread of AIDS. The major stimulus for behavior change in this area has been education, most of which has been generally, rather than individually targeted. While sexual behavior change because of the AIDS epidemic appears to be occurring in some populations (e.g. homosexual men), it is not known how less studied but equally susceptible subgroups, particularly adolescents, are responding to the AIDS epidemic. The long term objective of this study will be to contribute to the knowledge base required for nurses to develop individualized, effective educational programs regarding safe sexual behavior (SSB) for adolescents. The study will have several primary aims: 1) to determine factors associated with SSB among college freshmen; 2) to develop a mathematical model using the dependent and independent variables for predicting SSB in college freshmen; and 3) to establish reliability and construct validity for the Modified AIDS Information Survey (MAIS) and the Safe Sexual Behavior Survey (SSBS). To accomplish the aims the researchers will first determine construct validity and reliability of the MAIS and SSBS using factor analysis, Kuder-Richardson's 20 and Cronbach alpha as appropriate. Following this, the investigators will survey college students in three private universities within a major biracial metropolitan area. An anticipated 760 subjects will complete three instruments: the MAIS, the SSBS, the FTP inventory; and a demographic data form. Standard statistical procedures using SPSS-X will be employed to test the relationship between the variables. The Kolmogorov-Smernov test will be used to test the validity of the normality assumption; and standard regression techniques used to generate and test the mathematical model. Secondary aims of the research will be to explore differences in SSB between males and females, and among ethnic groups; and to test aspects of the Health Belief Model in adolescents, thereby providing direction for development of individualized educational strategies for the particular needs of the group.